Partial Credit

Jul 12, 2021

In advance, I'm pretty sure this issue has been plaguing Storyline for at least 8 years and still is not resolved. But here goes. Attached below is a 2-question quiz. Both questions are drag-and-drop. Each is drawn from its own bank of 5 questions. We want to be able to give partial credit where credit is due. And even if that continues to be impossible, ... Enjoy! It takes a while to construct these questions, mostly of interest to those who love pianos.

14 Replies
Maria Costa-Stienstra

Hi, Jason, and welcome to E-Learning Heroes! ✨

Thank you for sharing your .story file! I really enjoyed seeing how you set up those questions in very interesting ways! 

I'll leave it to our community members to offer you suggestions on accomplishing what you're looking to do, but I wanted to share a tutorial from our YouTube channel. It was created using an older version of Storyline, but the principles still apply:

I followed the tutorial with one of your questions using Storyline 360, and I'm attaching it here as a suggestion.

Let me know if it works!

Jason Kanter

I think this direction has promise, but I foresee some problems and would very much like your continued advice.
(1.) Dragging the labels to the diagram - For the most part, this type of question will have maybe 10 drops. A complication comes because user may drop drag1 onto target1 and get a point, and continue with drag2 through drag10 correctly, and then change his/her mind and move drag1 onto target10 and drag10 onto target1. So the variable "answers" would have to subtract a point when the correct drag is removed. That complicates the trigger design, no?
(2.) The same question design will be reused for all ten (or more) of these drag-n-drop questions in the given quiz. So the variable "answers" needs to be reset to zero when the slide is activated, no? Or do there need to be ten (or more) variables, answers1, answers2 and so on?

Jason Kanter

...also, the snap-to-center behavior is not working when the question is converted to MC. That is, it snaps properly if I drop the piano on the edge or corner of the target, but if I drag it into near-the-middle, it does not snap and does not register as correct even when I click Submit. Also, the "only-one-item-per-target" rule doesn't seem to work. But these issues are trivial compared to inability to subtract if learner changes her mind and moves a correct drag out of place. Are we hosed?

Maria Costa-Stienstra

Hi, Jason.

That's a great call! I reviewed the question, and here's an alternative:

  • Only add the values to the variable answers when the user clicks Submit. That way, they can change their minds as many times as they want, but the value will only be set when the Submit button is clicked.
  • We'll do that by adding a trigger that checks if the pictures are in the drop correct state.
    • For example: Add value 1 to answers when the user clicks submit if the state of Picture 2 = Drop Correct.
  • We'll add triggers to change the states to drop correct when they are dropped in the right target.
    • For example: Change state of Picture 2 to drop correct when the user drops picture 2 on ff1.

I didn't notice the snap-to-center issue. It seems to be snapping regardless of how far from the center of the target I drop it. Let me know if you see something different than below:

Screen Recording 2021-07-14 at 04.23.28 PM

I am attaching the edited version of the .story file if you want to test it out.

Jason Kanter

Very helpful. I'll work on building one of the more complex DnDs and will post the next iteration back here.

I'm attempting to attach a GIF of what happens when I place the drag into the target vs. placing it overlapping the target edge.

What software do you use to create an animated GIF and paste it into your response?

Maria Costa-Stienstra

Hey, Jason.

I didn't experience what you shared above, even when copying the exact movements from your animated GIF, but I'm happy you were able to solve it!

You also asked about creating animated GIFs that you can paste into the forum. Some of the tools we use are CloudApp and SnagIt.

 I hope this helps!

Jason Kanter

I've resolved a couple of problems and now this is working pretty well, see attachment. There is still the issue that multiple drags can be snapped to the same target - in a real DnD, we have the option to forbid multiple drags on the same target. That would be nice but is probably too complex? And it would be nice to shuffle the drag items if possible. 

Summary: The slide has a fundament of an MC question hidden behind a white rectangle; a screenshot of a Word paragraph with equally-sized gaps in it; six "target" white rectangles with a light blue outline, all the same size, placed on top of the paragraph; and six red words that are the drag objects along with another four "red herrings" that can be dragged but do nothing; the Triggers contain ten Object triggers and a bunch of Scoring triggers; when you click Submit, the Feedback tells you how many points you got.

I'm happy. I guess a lot of people will want to use this type of question.

Jason Kanter

The major assessment I'm working on has three types of questions where I want to award partial credit:

  1. Diagrams where the user has to drag the right labels to the correct parts. These can contain typically 6-8 drags, but I've done one that had 20. There are draggable red herrings as well.
  2. Gapped paragraphs where the user has to drag the right words to the correct gaps. Typically the paragraph of 60-70 words with 6 words missing, and there are an additional 3-4 red herring words.
  3. Matching drag-and-drop, where the user has to drag the chunk-of-words to interlock with the matching chunk-of-words, for example drag the definition to the word. These typically have 6-8 matching pairs.

For all of these I want to award points for however many the user gets correct.

Earlier in this thread I've learned how to accomplish partial scoring for types 1 and 2. See "2021Jul19-ParaDrag.story" to see how it's done. I put instructions (to myself really) in the Notes field.

Now I want to address Type 3, the Matching Drag-and-Drop. Certainly I can just build it from scratch, using the same MC underlying architecture, and assigning drags and targets to look just like the default Matching question. But is there a simpler way, using "add 1 to score when the state of Matching item 1 is Drop Correct"? Will that work?

Jason Kanter

(continuing to ponder) .. The problem is that the Matching question does not allow for scoring by choice. If it did, then we could arrange to select the appropriate score and have that added to the quiz total. But noooo. We can create a variable called "score" and add to it based on the number of correct drags, but the value of "score" will not be incorporated into %Results.Score%. Can we workaround, or are we limited to building something on top of a MC substructure?

Lauren Connelly

Hi Jason!

Thanks for sharing examples and providing additional details! I know this will help our community members out who are interested in partial credit. We don't currently have a workaround, so you are limited to using Multiple Choice.

If this feature makes it on our feature roadmap, we'll be sure to update you in this discussion.

In the meantime, I'm looking forward to seeing how community members have applied partial credit in their courses! I hope they'll chime in.

Jason Kanter

Actually I tried creating a Matching Drag on top of a MC, and it doesn't work, because the Matching question relies on other drags moving out of the way automatically when you drag something anywhere. Clearly it is programmable somehow, but I don't have the time. But it sure would be good to score those by choice instead of by question. 

Jason Kanter

I am requesting urgent help. I left a long voice message on your support line. The attached story gets hung on the Results slide. There are several questions using the drag-n-drop vs MC workaround, and maybe that's causing Results Slide hang? I have based this results slide on a Blank rather than a Graded. We have to publish today, and that won't happen.