Stanislav Duris

Stanislav Duris on Exponential increase 6 years, 9 months ago

Saved a checkpoint:

This question should be good to go. If not, the only thing I can think of as an advice is maybe you can adjust the first part of advice so that if years = 2, the latex does not display the 2 while calculating the root. This would be done via conditional visibility but I don't think doing this is necessarily essential. Maybe playing around with numbers would change things (the rounding could result in slightly different correct answers), so make sure that slight errors are allowed if that happens.

Stanislav Duris on Cumulative percent decrease 6 years, 9 months ago

Saved a checkpoint:

I hope this question won't need more huge adjustments, especially in part b, cause this part's advice was quite tricky to write and the answer is a bit tricky to calculate. I'll try to describe how the answer to part b) works:

1. At first, a random number from 6 to 9 is picked (testn).
2. Then pricee1 is calculated, which is the smartphone's value after testn months.
3. Threshold is calculated, which is the crucial number in the question. This is the first multiple of ten higher than pricee1.
4. We would normally be finished, but the numbers and percentages are not big enough so sometimes the decrease in the smartphone's value is less than 10, which means the smarphone's value one month before testn months may also be below threshold.
5. We calculate the smartphone's value after testn-1 months, this is pricee2.
6. Finally, the correct answer is the variable n. This is either testn-1, if pricee2 < threshold, or testn, if pricee2 > threshold.

Right now, pricee3 (the value after testn-2 months) never falls below the threshold (thank god) so it is not used in the question at all. However, if there'll be some adjustments to the price or the percentages in the question, this may not be true anymore. Make sure it is still ALWAYS below the threshold and if not, adjust the variable "n" so sometimes it can also be testn-2. I hope this is not too confusing and if it is, I hope no one has to deal with this anymore.

Stanislav Duris on Calculate the original price before a decrease 6 years, 9 months ago

Saved a checkpoint:

I made this question after Vicky suggested making a question where the student is asked to find the original 100% value given a value of for example 85%, because many students seem to struggle with this type of a question. If it needs any adjustments, please make sure the focus is still on finding the original 100%.

Stanislav Duris on Calculate a student discount 6 years, 9 months ago

Saved a checkpoint:

I hope this question is ready to use now. If not, I don't think I can provide any helpful comments for the person who picks it up, but I think (or hope) it would only need very slight adjustments.

Stanislav Duris on Find bounds for distance and time spent running, given imprecise measurements 6 years, 9 months ago

Saved a checkpoint:

I am satisfied with the question as it is right now, but maybe it is going to need a bit more work by someone else. This question is supposed to focus on multiplication and division of variables that have limits of accuracy and whether students understand how approximation and upper/lower bounds work. Maybe the conclusion in part b) needs a separate part or some other adjustment, it may feel a bit messy right now. The variables are constructed so the upper bound is always above the time asked in the question.

Stanislav Duris on Rounding and estimating calculations 2 6 years, 9 months ago

Saved a checkpoint:

Vicky suggested making this question and these parts because she thought the other question about estimation was not applicable to real life estimation enough. If this question still needs some stuff to do and someone else picks it up after I leave, you can maybe change some of the scenarios or adjust the question so there is no specific rounding students need to do and they can just choose one they want as Christian suggested. I sadly do not have time for this.

Stanislav Duris on Rounding and estimating calculations 2 6 years, 9 months ago

Gave some feedback: Needs to be tested

Stanislav Duris commented on Calculate the areas of polygons 6 years, 9 months ago

Hey Aiden,

I haven't noticed Chris has already made this suggestion when I started fixing part a) for you so you could see how to fix other parts. I've done it, so part a) should now work. Don't forget to put everything in one line before you do the conditional visibility.

There was one problem I came across. Sometimes, even when wh00 = wh00dp was true, Numbas thought it was false. When I made wh00 round to 2 decimal places (so practically stay the same as it already was), this problem seems to have vanished so I advise doing the same for other parts. It is some kind of an error.

Stanislav Duris on The probability of an event not happening - five friends play mini golf 6 years, 9 months ago

Gave some feedback: Has some problems

Stanislav Duris commented on The probability of an event not happening - five friends play mini golf 6 years, 9 months ago

This is a nice question, but I have some comments.

• While the parts are very clear, I feel like it looks a bit daunting that the statement is almost as long as the whole question. I would personally put some of the statement in the advice (so the first time they do the question you test them on this and the next time they redo it, they'll know all this helpful information) but it is up to you.
• Part a) is a good simple question. Just a tiny unnecessary comment, your Hoop variable has a description of "Denominator for probability that your car runs out of fuel before you get to Leeds", I assume that this was how you first wrote the question and changed it to a different situation afterwards forgetting you have to fix your variable description.
• Also, you used random(5..9 #1) in your variable, a step size of 1 is the preset setting so the #1 is unnecessary there. It isn't wrong at all and you should keep it there and maybe you realise this, I just thought to let you know in case you need to use random integers in future so the job is a bit easier for you.
• Also, when Hoop = 7, Numbas does not accept the decimal form of the answer my calculator gives. I don't know if this is a big deal, but maybe some students would try to put decimal forms instead of fractions in. Maybe you could say in the question put your answer as a fraction?
• Part b) - I am not 100% convinced about pass the parcel being the right game for this question, but it is not a big deal.
• I feel like it would be better to swap parts i) and ii), because to get the answer to part i), you already had to calculate the answer to part ii). Maybe the order is like it is right now because you were aiming to test whether they realise this so if that is the case, that's fine.
• The advice for part b)ii) is maybe a bit too complicated considering how simple it is in reality, especially when one of the methods you use is the same as in part a). I think putting the second method first would maybe be better.
• If you swap parts i) and ii): in advice for the new part ii) (calculate the probability of Dexter winning), you could use P(E) = 1 - P(E'), a slightly different turn on the equation for complementary probabilities. This would make part b)ii) a bit different from part a).
• You use latex for $P(.)$ which is fine but I would personally use $\mathrm{P}(.)$ as it looks nicer in my opinion. This is coded as $\mathrm{P}(.)$.

It is a good question as it is right now. This is all just my personal opinion of what you could change so it is up to you to decide which things are better now and which things would be better changed.