Tincan statements differ for Flash and HTML5

Dec 20, 2016

Darren Hepe
By Darren Hepe

Greets all

I am developing a template quiz in Storyline 2. It is a question bank with 5 multiple choice questions. Navigation between the questions is with the previous and next buttons, with only the last one having a submit button - the design calls for students being able to navigate back to questions and answer them differently if they wish, only being locked into answers for the entire quiz when they push the submit button on the final question. The student upon re-entering the quiz sees all their previous answers pre-checked, and a green or red box (with an icon) around the answer if they got the question right or wrong respectively on the previous attempt. I have a special setup on the final slide to exit the course, but re-start the quiz on the first slide when the quiz is re-launched with the state object still loaded, so resume data is retained. The question bank is linked to the results slide, which is tracked for Tincan reporting.

This almost works how I want it to, except for Tincan API statements. The issue is that at the end of the quiz when flash is enabled in the browser, there are six important Tincan API statements sent to the LRS - one for the quiz overall (passed or failed, and what score), and one for each of the five individual questions in the quiz (the question, responses, and what the student chose, and whether or not they got it right).

When flash is enabled in the browser this is exactly what I get. But when flash is disabled and the quiz is launched in HTML5, only the overall result slide Tincan API statement is sent to the LRS. The five individual Quiz question Tincan API statements are not sent at all.

I have tested this in both SCORM cloud, and the Learning Locker LRS we are using - the result is exactly the same.

Does anyone know how to overcome this issue?

(p.s. I have successfully used manual crafted TincanJS statements to the LRS in javascript triggers, using Oauth authentication, but since other developers need to use this template, I want to avoid this as much as possible if Storyline 2 default reporting can work properly in HTML5)

(p.p.s I forgot to mention I do not use submit interaction triggers in any of the questions. Doing so would lock the question answer in which is against the design spec - students must be able to navigate previous and next between questions, and all of the questions need to be locked in all at once when the hit submit on the last question, and got to the result slide. This seems to work fine in Flash, but not HTML5 with missing Tincan API statements. Almost all other statements - such as experienced etc work fine in HTML5)

6 Replies
Ashley Terwilliger-Pollard
over 7 years ago12/20/16 at 3:22 pm (UTC)

Hi Darren,

Thanks for reaching out here and crafting such a well written and concise account of the issue you've experienced. It's incredibly helpful! 

I would like to have you work with our Support Engineers so that we can do a bit more investigation with you, as we'll also be curious to find out more about your particular file so they'll share directions on how you can send it along. It'll also allow us to make sure that it's reproducible in a new file as well.

I'm going to open up a case on your behalf so you'll hear from our team shortly. 

Darren Hepe
over 7 years ago12/21/16 at 1:19 am (UTC)

Hi Ashley

I have sent my project file to one of your Support Engineers, thank you to your team for a fast response.

I have also noticed that if I do use submit interaction trigger in the individual quiz questions (which I want to avoid) the Tincan API statements sent with flash enabled differ from what is sent when flash is disabled (and fallback to HTML5). Here is the difference:

With Flash:

"object": {
"objectType": "Activity",
"id": "http://CHCCOM005_example_Assess_01_Quiz_01_beta/5uH5OvEw2Xq/6JXom5SvZ6B",
"definition": {
"description": {
"und": "Which of the following is relevant in relation to Duty of Care?"
},
.
.
"name": {
"und": "Which of the following is relevant in relation to Duty of Care?"
}

With HTML5:

"object": {
"objectType": "Activity",
"id": "http://CHCCOM005_example_Assess_01_Quiz_01_beta/5uH5OvEw2Xq/6JXom5SvZ6B",
"definition": {
"name": {
"und": "CHCCOM005_example_AS01_Quiz01_Q2"
},
"description": {
"und": "CHCCOM005_example_AS01_Quiz01_Q2"
},

---

As can clearly be seen above, the name and description of the object (the activity) is reported to the LRS differently, Flash reports the quiz question text for the name and description. HTML5 reports the slide name for the name and description.

When viewed in the LRS they appear differently as follows:

Flash:

TestDarren TestHepe answered Which of the following is relevant in relation to Duty of Care?

HTML5:

TestDarren TestHepe answered CHCCOM005_example_AS01_Quiz01_Q2

 

This creates several problems. The HTML5 statement does not contain the actual quiz question text anywhere, making it difficult to know what question the student was presented with (especially if later on a new version overwrites an older one with new text). The other issue is the Flash statement does not contain the slide name, which is what I would want to report on for the quiz question, since that will not change regardless of future question updates, and I can use it to compare student responses for that question. I suppose I can use the object ID for comparison, but is there any way to see what the object ID will be for a slide or its components before publishing and checking the LRS after running it?

Ashley Terwilliger-Pollard
over 7 years ago12/21/16 at 3:08 pm (UTC)

Thanks  Darren - this is all really helpful information to have! I see that Robert is also testing the file currently, so I'll be sure he has this additional information as well. 

As for the object ID, there isn't a way to see it prior to publish - it's likely only visible within the LMS or the debug data captured here. 

This discussion is closed. You can start a new discussion or contact Articulate Support.