This is just a really quick post about me not putting 2 and 2 together as soon as I’d like, but I need to write this down so I don’t forget again. I’m categorizing this as #InArticulateeLearningHero, though it is not an official challenge from the elearning heroes group.
Over the past few years there have been courses that include some type of personality test activity. These might not be Meyers-Briggs officially, or actual personality tests, but they fit the format of:
- identify which of the following statements you fall into
- add a point for certain items
- subtract a point for other items
- tally your score and read the results
H5P does have a personality quiz content type, however it doesn’t work in quite the same way many of the paper based quizzes work. For the H5P personality quiz, you create items and then each answer is associated with a personality. At the end of the quiz, which ever personality has the most points (answers you submitted associated with it) then that’s your personality. It’s done as first past the post, take it or leave it. I have created activities which used this content type, but it was a real work around to get it to work.
Here is one possible alternative, which allows you to add and subtract points for statements. It doesn’t give you as much flexibility as some other tools, such as those that associate points to different levels on a likert scale, but it may be useful for some applications. Originally I tried doing this with a question set. I thought I could make a series of true/false questions, give points for correct answers, and subtract points for incorrect answers. This meant that each statement would be a question stem, followed by the ability to click true or false.
What I came to eventually was flipping the question on its head. So instead, not it is a single MCQ and each statement is an alternative. Each alternative that results in a point is checked as correct, and for each one I need a negative point for is left as incorrect. By default, a “multiple answer” style MCQ in H5P will subtract a point for incorrect answers. This is done to prevent just always checking every alternative in order to achieve a top score. Then, the feedback is divided into specific ranges to reflect the final personality score. The accidental benefit of this feature is that I can now make a style of personality quiz using this content type.
Try out the example:
Photo by Brett Jordan on Unsplash
2 responses to “Personality type tests with H5P”
I had always ignore the Personality Quiz type until I saw this clever example https://h5pstudio.ecampusontario.ca/content/698
I am not quite sure what you mean as “first past the post, take it or leave it” – that meaning that responses only build toward a tally, that there are no detractors? I liked what you did in your example, but isn’t that just making use of the feedback based on percentage? What I like about the multiple question ones is that there are several streams of input, and the detractors are not quite obvious.
And even as a question set, it becomes really just an overall performance rating, not anything that builds on the path through. Now I am thinking this might be a thing for branching scenario??
Always appreciating your experimentation here, JR!
Now that is an interesting example of the personality quiz from eCampus!
Unless there has been an update, the first past the post means that which ever of the “personalities” gets the highest tally is what is presented to the user. So if you answer two questions in a way that points to personality 1, and one question that points to personality 2, and 3 that point to personality 3, you would be presented with personality 3 at the end. The questions that point at the other programmed personalities then have no bearing on the results.
What the MCQ method allows for is a bit of a sliding scale, using the ranges you might be able to set up something a touch more complicated. The feedback and points end up being the limitation, as you can’t have one item worth more value, or provide anything more than plain text in the feedback.
Of course the branching scenario or presentation give you the most flexbility to create something with a lot of different options, but for this particular use case I had recently we were looking at getting as close to the paper version as possible and then tie those results into the rest of the module in the learning materials text in the LMS.
Thanks for the comments and feedback!