Auto Multiple Choice - A more comprehensive Scoring Strategy

A more comprehensive Scoring Strategy

Suppose a question/multi-question has options 1, 2, 3, ....n.
If option (i) is ticked/marked/bubbled, then a variable (m_i) equals 1 and a variable (M_i) equals 0 &
if option (i) is not ticked/marked/bubbled, then the variable (m_i) equals 0 and the variable (M_i) equals 1.

The score I would like to assign is (a_1 m_1 + a_2 m_2 + .....) + (b_1 M_1 + b_2 M_2 + ....).
(for some numbers a_1, a_2, ...., b_1, b_2, .... of my choice)

But I don't know how to implement this universal construct.

It would have been nice if for every question, I could say:

\begin{question}
          ......
    \begin{choices}
       \choice{ .......}\scoring{a=#1, b=#1} % a=#1: #1 is a_1 above   ; b=#1: #1 is b_1 above.
       \choice{........}\scoring{a=#2, b=#2} % similar to above

......
    \end{choices}
    \end{question}

This way of allowing choices seems more universal than the traditional \correctchoice{} and \wrongchoice{}
constructs.

Replies (1)

RE: A more comprehensive Scoring Strategy - Added by Alexis Bienvenüe over 11 years ago

Use of \correctchoice and \wrongchoice is useful to build the solution document, and to annotate students' completed answer sheet.
If you don't need these things, you can use something like that:

  \begin{questionmult}{ID}\scoring{MAX=#1} % use #1 = max(a_1,b_1)+max(a_2+b_2)+...+max(a_i,b_i)
    ...
    \begin{choices}
      \wrongchoice{...}\scoring{m=#1,b=#2} % a=#1: #1 is a_1 above   ; b=#1: #1 is b_1 above.
      ...
    \end{choices}
  \end{questionmult}

Using \wrongchoice for all choices, the score m is used when the answer is "not correct", which means that the box is ticked, and b is used when the answer is "correct", which means that the box is not ticked.
However, using this method, AMC won't compute the correct maximum score, as it computes it applying the scoring strategy to the "perfect" answer (boxes \correctchoice ticked, and \wrongchoice not ticked). So you need to add a \scoring{MAX=...} for each question.
To solve this problem, you can consider defining a LaTeX command \choice{text}{a}{b} that compares a and b and calls \correctchoice if a>b and \wrongchoice if a<b, with adequate scoring.

(1-1/1)

Login	Password