Uploaded image for project: 'Moodle'
  1. Moodle
  2. MDL-48042

Reset Calculated qtype comments on datasets to their initial coding

    XMLWordPrintable

    Details

    • Testing Instructions:
      Hide
      • Go to any question bank
      • Create a new 'Calculated' question
      • Set the following fields:
        question name: Test MDL-48042
        question text: Test MDL-48042
        answer 1 formula = ( {x}*0.001)*{y}
        grade: 100%
        answer display: 3 significant figures
        * click save changes
        * click next page
        * Set the following fields:
        wild card {x}

        : 0.240
        range min: 0.15
        range max: 0.45
        decimal places: 3
        distribution: uniform

      wild card

      {y}

      : 84.2
      range min: 20
      range max: 100
      decimal places: 1
      distribution: uniform

      Tolerance ±: 0

      • Correct answer shows*: 3
        format: significant figures
      • click update the answers tolerance parameters

      Expected: ERROR Correct answer : 0.020 outside limits of true value
      Min: 0.020208 — Max: 0.020208

      Modify * Correct answer shows* to 7

      Expected: Correct answer : 0.020208 inside limits of true value
      Min: 0.020208 — Max: 0.020208

      This is a copy of the test done when testing MDL-47224 although here the expected result is different.
      TO complete the test, create 100 set(s) of wild card(s) values.
      There should be no error ( I tested 10 times).
      Preview using the preview button.
      Use Fill in correct response and submit.
      You should obtain a 100% grade like the Test_MDL-48042.jpg

      Notice that setting Tolerance to 0 is tricky as it implies that the 2 values are really equivalent.
      In this specific test the number 7 was found by trial and error and its related to the fact that the answer equation and the range chosen for the variables does not give a response with a greater number of decimals.

      Show
      Go to any question bank Create a new 'Calculated' question Set the following fields: question name: Test MDL-48042 question text: Test MDL-48042 answer 1 formula = ( {x}*0.001)*{y} grade: 100% answer display: 3 significant figures * click save changes * click next page * Set the following fields: wild card {x} : 0.240 range min: 0.15 range max: 0.45 decimal places: 3 distribution: uniform wild card {y} : 84.2 range min: 20 range max: 100 decimal places: 1 distribution: uniform Tolerance ± : 0 Correct answer shows*: 3 format: significant figures click update the answers tolerance parameters Expected: ERROR Correct answer : 0.020 outside limits of true value Min: 0.020208 — Max: 0.020208 Modify * Correct answer shows* to 7 Expected: Correct answer : 0.020208 inside limits of true value Min: 0.020208 — Max: 0.020208 This is a copy of the test done when testing MDL-47224 although here the expected result is different. TO complete the test, create 100 set(s) of wild card(s) values. There should be no error ( I tested 10 times). Preview using the preview button. Use Fill in correct response and submit. You should obtain a 100% grade like the Test_ MDL-48042 .jpg Notice that setting Tolerance to 0 is tricky as it implies that the 2 values are really equivalent. In this specific test the number 7 was found by trial and error and its related to the fact that the answer equation and the range chosen for the variables does not give a response with a greater number of decimals.
    • Affected Branches:
      MOODLE_26_STABLE, MOODLE_27_STABLE, MOODLE_28_STABLE
    • Fixed Branches:
      MOODLE_26_STABLE, MOODLE_27_STABLE, MOODLE_28_STABLE
    • Pull from Repository:
    • Pull Master Branch:
      MDL-48042_wip

      Description

      MDL-47224 modify the way datasets handle the handling of $answer->min and $answer->min are calculated which could result in setting a good response with a 0 grade.
      see MDL-47224 for more details

        Attachments

          Issue Links

            Activity

              People

              Assignee:
              ppichet Pierre Pichet
              Reporter:
              ppichet Pierre Pichet
              Peer reviewer:
              Tim Hunt
              Integrator:
              Dan Poltawski
              Tester:
              David Monllaó
              Participants:
              Component watchers:
              Tim Hunt, Andrew Nicols, Jun Pataleta, Michael Hawkins, Shamim Rezaie, Simey Lameze
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved:
                Fix Release Date:
                12/Jan/15