Trial mark sequence (Formulas question)

Trial mark sequence (Formulas question)

by Dominique Bauer -
Number of replies: 27
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Jean-Michel,  Bernat,

Do you know if the Trial mark sequence works? (see https://docs.moodle.org/33/en/Formulas:_Answers_and_marking#Trial_mark_sequence)

Average of ratings: -
In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

No, I removed this feature back when I ported formulas to the quiz engine 2 for Moodle 2.1.

Better to remove this section from the docs.

If I remember well my thinking was that it was more a feature related to behaviours than to question type.

Of course if somebody one day write such a behaviour this would need to add a field in the question type to store the sequence, but the working would be better done in the behaviour.

Do you think this would be an interesting feature ? At that time my impression was that very few teachers would use it, but maybe I was wrong.

My priority was also to have it working with other behaviours because I planned to use it with the interactive with multiple tries behaviour.

I must also admit that my first goal was to have the question type working and that when I made that decision I was far from that goal and thinking that maybe I would never reach it ! So removing minor features was not a real problem.

I will see if I find other features in the docs that needs removing.

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

The section (https://docs.moodle.org/33/en/Formulas:_Answer_boxes) needs some adjustments:

The styles in styles.css have changed names since some time now see

https://github.com/jmvedrine/moodle-qtype_formulas/blob/master/styles.css#L44

And I just see that the formulas_algebraic_formula_unit for Algebraic answers with combined unit style is missing (probably it was never done !) I will add it

Here is a screenshot of the result for answer and answer with separate unit

Formulas question answer boxes width


In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

In the https://docs.moodle.org/33/en/Formulas:_Answer_boxes#Quantity_and_unit_boxes_side_by_side_or_merged maybe it would be good to say that the unit bow can only be merged with the last quantity box of a part (if the part has several quantity boxes of course)

For instance {_0}{u} {_1} is invalid

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Bernat Martinez -

Hi Dominique, 

Congratulations for your hard work in creating the documentation for formulas question. 

Just one comment, at CORRECT ANSWER section I would stress that the variables for algebraic formula answer should be written at the part local variables.

I have seen many teachers writing them at random variables, it works also but when the correct answer is shown it shows the formula with the variable substitute by a number.

In reply to Bernat Martinez

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

If I am not mistaken they can be defined as Global variables or Part local variables (it depends what scope you want to give to these variables, for instance if you plan to use them in several parts, you need to define them as global variables)

But you are right if you define them as random variables, a number is chosen and displayed.

Dominique's documentation is right in stating that for formulas question an algebraic variable is in fact a set of numbers, the numbers that will be used to test all expressions involving this variable. The numbers in the set are the points where evaluation will take place.

So defining it as a random variable is wrong.

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

In the https://docs.moodle.org/33/en/Formulas:_Numerical_functions you can add the perm, comb, gcd and lcm functions useful for probability and arithmetic questions. The gcd and lcm need no further explanations.

comb(n, r) is defined as \( \frac{n!}{r!(n-r)!} \)

perm(n, r) is defined as  \( \frac{n!}{(n-r)!} \)

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

Maybe you should add somewhere that in adaptive mode (in fact but that is not important for users when you select Adaptive behaviour, it's Tim Hunt Adaptive with multiparts behaviour that is used) when you click on the Check button :

  • No penalty is applied for parts that have no answer entered
  • No penalty is applied for an answer to a part that has already be given in a previous check (Students aren't penalized for doing the same mistake again)
So:

  • Students can try to answer one part after the other without loosing any mark. 
  • And if they give a wrong answer for one part, they can let it as it is and check new answers for other parts.
This was very important for me when I designed the version for Moodle 2.1, because Hon Wai Lau original version for Moodle 1.9 and 2.0 had a submit button for each part. But given the way the adaptive mode is working in Formulas, I saw no real reason to keep all those submit buttons. 

And that eliminated for me the need to do a special behaviour with multiple submit buttons. wink

I have now finished reviewing all the docs for the Formulas question and let me say that I am really, really impressed by the quality.

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Jean-Michel,  Bernat,

Thank you for your comments. Here are a few points:

  • I only edited Hon Wai's text by changing the presentation, making some updates and adding examples.
  • It's a long-term job that is not finished yet, for which I work only part-time. There is still a lot of work to be done. For example, I started adding Hon Wai's thirty original examples from Google Code Achive.
  • I am also working on the French version of the documentation.
  • I have also defined the corresponding sections in Spanish in order to facilitate the work. Puedo comenzar la traducción, pero tendrá que ser revisada por alguien cuya lengua materna sea el español.
  • For now, the examples are hosted on a dedicated Google Cloud Platform site. The Spanish package must be installed, accounts must be opened to allow contributions, etc.
  • I will try to make the changes based on your posts, but you can also do them directly.
Average of ratings: Useful (1)
In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

If I remember well, 2 of the 13 examples from the Hon Wai Lau file have problems. I will try to find the mail where I reported the problems to him a long time ago and explain the needed corrections. The "Possible question type - quadratic equation" is wrong from a mathematical point of view but works as expected.

I think that the file included in my version is corrected but I need to check to be sure.

I know I can make the changes but this morning I wanted to review the docs using the free time I had, and it was easier for me to have 2 browsers windows open, one to read the docs and one to post the changes. I will use this thread as the basis to make the changes.

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

Oops it seems that I made a mistake and in some parts of the code comb is called ncr and perm is called npr

I absolutely need to correct that mistake before releasing and before these fonctions are added to the documentation.

Wich version do you prefer ?

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Jean Michel,

You are the professor of statistics, hence you know better than us.

I think that nCr and nPr are preferable because they are commonly used and unambiguous. Indeed, they refer to the number of combinations and number of permutations. comb and perm are not as common and both also refer to other functions in other disciplines.

In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

Well I was teaching statistics smile

I also think that nCr and nPr are better but I will use lowercase only for consistency with other functions : ncr, npr

A good reason to do so is that these are the names used on scientific pocket calculators.

Average of ratings: Useful (1)
In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

I searched my old mails and back in 2011 I wrote to Hon Wai lau that 2 questions in his example file had problems

question 3.2: Grading - multiple criteria
text says: Two grading criteria are defined here, each of them worth half mark of the question.<br>
<br>
but contrary to what is said if you enter an answer that only verify one criteria, you are not awarded any mark
And also
question 5.5: Possible question type - quadratic equation. 

The correct answers given in the question are not the good ones

The first problem is easy to correct (maybe it is already corrected in the version of the file that you use)

Part text: Give a number x that is a multiple of {d} and {a} < x < {b}.

Grading variables: criterion1 = _0 % 7 == 0; criterion2 = a < _0 && _0 < a+10;

Correctness: criterion1 && criterion2

Of course that doesn't work, correctness should be (criterion1 + criterion2)/2

The second problem needs more work, but I have a far better version of a question about finding the roots of a quadratic equation, so I will include it in the docs (and now that I have integrated the poly function contributed by PeTeL-Weizmann the display is even better !)

In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

I see no problem in the documentation. I was fearing that you had used the original file on Hon Wai Lau website that still have these problems.

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Mary Cooch -
Picture of Documentation writers Picture of Moodle HQ Picture of Particularly helpful Moodlers Picture of Testers Picture of Translators

Hello both. Good to see this documentation smile Would I  be correct to assume that this will also apply to Moodle 3.4? Because if so then feel free to add it to the 3.4 documentation, or if you are too busy, I can do so .

In reply to Mary Cooch

Re: Trial mark sequence (Formulas question)

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Mary,

Thank you for your message. It deals with a point that I checked with Helen Foster, but that will need to be reviewed as new versions of Moodle keep coming.

The documentation currently on docs 3.3 also applies to recent versions of Moodle (2.0 and above) and as well as Moodle 3.4.

The documentation consists mainly of the original documentation by Hon Wai Lau, which I edited and modified as needed, and to which I added other examples and explanations. The documentation takes about twenty pages and contains many screenshots and has many links to live examples (Play it buttons), and it includes its own table of contents. Categories are defined and dead links are redirected.

I am also translating the documentation into French and into Spanish, and although these translations are not completed, they are well underway.

All of this represents a significant amount of work, and the work is still ongoing.

I found it unnecessary to reproduce all the docs 3.3 contents in docs 3.4 because, since the Formulas question is a plugin, there is essentially no change between Moodle 3.3 and 3.4. Reproducing the plugin documentation for the different Moodle versions is a huge job in itself, that would bring little, and it would be a major source of mistakes and confusion.

I discussed this with Helen Foster and she agreed that it would be appropriate to have only one version of the documentation. I quote her: "I think it's fine having the Formulas question type documentation in the 3.3 wiki and adding a message to other docs wikis, as you've done for the 3.4 docs."

If we look at the French site, we have https://docs.moodle.org/3x/fr/, that is to say a common documentation for all versions of Moodle 3. In Spanish, we have https://docs.moodle.org/all/es/, a common documentation for all versions of Moodle. In the same way, for the Formulas plugin, it would be nice to have a place to put the documentation common to several versions of Moodle, and not have to reproduce it for each version.

Currently in docs 4.4, I have placed a link that automatically redirects to docs 3.3 where one can read:

     "As there is no or little difference in the Formulas question type plugin for recent versions of Moodle (2.0 and above), the documentation for the Formulas question type has been moved to one location (the present page). The documentation herein applies to all recent versions of Moodle including the current Moodle 3.4."

I think this is clear enough, but it would be even better if the documentation for the Formulas plugin could be placed somewhere that is independent of the Moodle versions. I think this place should be on MoodleDocs and not on an external site, even if the Formulas plugin is a third party plugin.

The examples to which the documentation refers directly are on an external site. They could have been hosted on MoodleCloud or MoodleSchool, except that the Formulas plugin can not be installed on these sites.

Your opinion would be appreciated.

In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Mary,

Given what I just said, it might be interesting to place the documentation for the Formulas plugin in a new wiki directory called either:

  • https://docs.moodle.org/3x/en/
  • https://docs.moodle.org/all/en/
  • https://docs.moodle.org/formulas/en/
  • https://docs.moodle.org/plugins/en/
  • https://docs.moodle.org/third-party_question_types/en/

I think https://docs.moodle.org/formulas/en/ could be the simplest and the best choice. In Spanish, this would be https://docs.moodle.org/fórmulas/es/, and in French https://docs.moodle.org/formules/fr/.

Do you think it would be appropriate and feasible?

In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Helen Foster -
Picture of Core developers Picture of Documentation writers Picture of Moodle HQ Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers Picture of Translators

Dominique, thanks for your suggestion, however a separate wiki for plugins would result in 'Moodle Docs for this page' links in Moodle no longer working.

In reply to Helen Foster

Re: Trial mark sequence (Formulas question)

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Hello Helen,

I thought it would be better to place the documentation on the Formulas question in a place independent of Moodle versions, since the plugin does not change at the same pace. This will avoid redundancies and risks of confusion even errors.

The documentation is at the address http://35.193.77.29/. Since this is a Moodle website, the documentation is easily accessible and transferable if needed.

I hope that you will find this initiative adequate and that it will encourage the use of Moodle with the Formulas question type.

In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Marcus Green -
Picture of Core developers Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers

The documentation for this is very impressive, thank you for your work Dominique

In reply to Marcus Green

Re: Trial mark sequence (Formulas question)

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Marcus,

Thank you for your compliment.

Since your message, I placed the documentation on an external site (http://35.193.77.29/) mainly because it is easier to manage that way. Nobody made any comment. I simply hope that it will be as convenient. Of course, suggestions, comments and contributions are welcome.

In reply to Dominique Bauer

Re: Trial mark sequence (Formulas question)

by Gary Marsh -

I think for the user, convenient access to current documentation is important.     Where it is located ... not so much.

The current documentation is a significant improvement to the original and congratulations to everyone on the formulae question team for a job well done!

G

In reply to Jean-Michel Védrine

Re: Trial mark sequence (Formulas question)

by Bernat Martinez -

Very good idea to integrate the POLY function contributed by PeTeL-Weizmann, I have already used and it displays nice polynomials.

In reply to Bernat Martinez

Re: Trial mark sequence (Formulas question)

by Jean-Michel Védrine -

Yes with it you can do an even better version of the quadratic equation question  \( a x^2 + b x + c = 0 \) where a can take negatives or positives values (of course not zero !) and the equation display nicely with no spurious +-

Ultimately you make 3 versions one for the case with 2 roots, one for the case with one root and one for the case with no roots in a question bank category  and you make a quiz with 3 questions chosen at random in this category and you have a real quiz about this subject that students can practice at will.