GR6 User Testing
Design & Implementation
Screenshot |
Design Description |
Implementation |
---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Evaluation
Testing Procedures
Three users were selected for testing. All three users have been on a trip that requires planning of what to pack - and have different problems with packing that could potentially be solved by this application. Two of the users explains that they go on a similar vacation every year. Two of the users describes that they are often unsure of what to bring when they go to somewhere they have never been to before. One user describes having forgotten to bring enough socks once.
Because this app is highly focused on efficiency (being able to quickly move and checkoff items), there are shortcuts that can be better understood by learning from other sources (eg: other people, or a short introduction video). In this case, learnability was sacrificed in this area in order to simplify the application, and to conserve the limited mobile screen space. The testing procedures were as follow:
- A briefing video was shown to the user that summarized the purpose of the application, and gave a few pointers to using the application. The video can be found here.
- Users started the application, and were given enough time to read the introduction.
- Users were given a series of index cards that contained a task (shown below), and were told to complete each one as specified. There were 11 tasks in total, each task targeting a different functionality of the application.
- Users received Skittles candy as compensation for taking time out of their busy schedules to test our application.
Raw Observations
These are observations that were recorded during the testing procedures. The document can be found here(opens a google document).
Summarized Observations
These are summaries of what was observed throughout the three users that tested PackIt.
No. |
User Task |
Summarized Observations |
Plans for Improvement |
---|---|---|---|
1 |
Create a new trip called "Alaska Vacation", by plane from June 1st - June 8th |
Users generally had no problem completing this task. The particular android phone we were using did not have a clear ENTER / DONE button on the built in keyboard, and one user was unsure of how to finish typing and close the window. |
None |
2 |
Pack 2 shirts into your backpack |
Two of the users intuitively dragged the items into the backpack, then realized it dragged all the shirts as opposed to just 2. One of the users attempted to drag all the items back, but was unable to do so as this function does not exist. Users that made the mistake were able to move the items back on the shelf after they had understood the difference between dragging and clicking, and easily corrected their errors |
Implement ability to DRAG items to different parts of the application, and not just to the backpack (eg, to the shelf, to the bubble that described what was in the backpack). Make the distinction more clear at the beginning - perhaps some popup TIP dialog. |
3 |
Pack all the socks |
One user tapped once, then realized it was possible to drag everything at once, and so did so appropriately. One user tapped multiple times in order to get all the socks into the backpack. One user dragged (most efficient method) |
Emphasize distinction between dragging and clicking (as mentioned above |
4 |
Unpack 1 shirt |
Users had no difficulty performing this task |
None |
5 |
Add 1 shirt to the shelf of things 'to pack' |
Users had no difficulty performing this task |
None |
6 |
Edit the shelf so that you won't be bringing any pants |
Users had no difficulty performing this task. One user attempted to click instead of dragging items to the trash can, causing a minor bug to happen |
Correct the bug that packs your item when you are in the 'shelf editing mode' |
7 |
Add 4 pairs of sunglasses to the list of things to pack |
One user was confused because the shelf did not change to the category when socks were being added, and so was unsure of whether or not socks were actually being added |
Modify program such that when anything is being added to the shelf, the shelf changes to the category of the item being added |
8 |
Add 1 fishing rod to the shelf of things 'to pack' |
Users had no difficulty performing this task, although when one user forgot to input the number, the original item that had been typed in disappeared. In addition, the screen becomes extremely cluttered when the keyboard appears |
Change the look of the screen when the keyboard appears, and make it so that what was typed does not get removed even if the entry is invalid |
9 |
Pack the fishing rod |
Users had no difficulty performing this task |
None |
10 |
Change your settings so you are now driving to Alaska |
Users had no difficulty performing this task |
None |
11 |
View the weather information for your trip |
Users had no difficulty performing this task |
None |
General Summary
It is clear that the distinction between drag and click is not clear (for moving items into the bag/out of the bag/onto the shelf). This can be corrected with several methods - having a tips page at the beginning that is shown to the user or showing a user a tooltip if they are detected to be clicking multiple times in order to move everything into the backpack. In addition, because of time constraints, the dragging model was not implemented the other way around (so moving items in the backpack back up to the shelf). This should be implemented as it would greatly help the consistency of this feature. As well, a minor issue arose when the keyboard appeared during the 'add non-existing item' feature. The screen became cramped as elements were overlapping, and an easy fix to this problem is to make the screen scrollable such that during this action the entire screen scrolls up to make room for the keyboard.
Reflections
What we learned
Over the course of the iterative design process, our team learned a great deal. The most important thing we learned was something very fundamental to UI, we were reinforced by the fact that the designer thinks very differently from the user. What seems so clear to the designer may be completely confusing the user. Because of this we had to try to think from the users standpoint, and make every aspect of our app as simple and clear as it could possibly be. Another thing we learned was the sometimes sacrifices need to be made for the overall goal of usability. In some parts of our app we needed to sacrifice learnability for efficiency. For example, one of our very first drafts of the app included three seperate tabs, one with a visual representation of the list of items to pack (aka backpack/shelf, ect) and one with a text representation of the list, which could be edited to add new items, ect. However, we realized that this would require the user to go back and forth between two modes many different times, which would be very slow and annoying. Although it would beeasier to see list of items to pack in two forms, it was a much better idea to keep it simple and efficient and only showcase one mode.
What we would change
If we were to do this project over again, we would do several things differently. We would prototype the main feature of the app more thoroughly. When changes needed to be made to the app after user testing, we would make sure to create/sketch a completely new draft that included these improvements, instead of just editing it a little at a time based on each new comment, as that allows for some sloppy errors in the long term. (For example, you change one thing but have to change it back based on other users opinions.) Anther thing we would change is to prototype a more realistic looking version of our application. The paper prototypes were much bigger than the screen size of any of our phones, and we felt that this allowed certain objects to look better on paper than they actually turned out to be on the phone. The final and most important thing we would do differently is to really listen to every comment a user makes. Sometimes it is easy to think that the designer is correct, or that the designer's implementation is good enough and that the user will "get it", but in the end the user is always correct. Some details the users mentioned were confusing ended up being a big problem when we finally created a working version of our application. More focus on these details during the testing phase would've helped us a lot for our final app.