After card sorting, the next usability test was paper prototyping. Paper prototyping is great because not a line of code has to be written to test your website. You draw (by hand or on a computer) what you want the pages of your website to look like and include the terminology tested in card sorting. You can even test features or functionality of your website. For example, if you have a tabbed search box on your home page, you can cut tabs in the top of index cards and you have the same look and functionality. You can very easily and quickly have 2 or 3 mock ups of any web page to test. Once your drawings are complete, develop tasks or questions for the user to do or answer using the paper prototype. Paper prototyping offers a lot of flexibility and valuable data about the terminology, functionality and organization of your website.
We conducted paper prototyping a bit later than we should have (there was already code written, major changes are much easier and less expensive to make if you do not have to rewrite code); however, we felt the insight and data would still be of great use. We opted to create mock ups for the home page, the first layer, and a few pages from the second and third layers. We recorded (without identifying information) the questions our users ask us in-person and via chat, email, and text. After analyzing the data, we had 18 categories of most frequently asked questions (e.g., hours, library account/renewing materials, finding books). From these categories, we developed 20 task based questions (e.g., Are there jobs available at the library?) but we only asked each test participant 10 questions as we did not want the test time to be too lengthy. Two librarians, one acting as the computer i.e., switching the paper as appropriate and one taking notes, conducted the study.
Participants were successful at
- finding hours (this is our number one question and with hours clearly displayed on the home page, it will be interesting to see if the number of times we are asked decreases),
- renewing books (through feedback, we discovered many people overlook the “my accounts” along the right so it is purple now),
- reserving a study room,
- finding information on a library event,
- accessing RefWorks (RefWorks is the application we offer to help you manage citation information),
- finding a book on course reserves,
- finding books (because they recognized ALICE but some were unsure about OhioLINK),
- finding a movie, and
- starting with ArticlesPlus to find an article (we did not mock up results from ArticlesPlus so it will be interesting when testing with the live website to see if the results make sense to users).
There were areas that participants struggled with completing the tasks, including
- locating the jobs in the libraries
- finding a phone number
- finding printing or software information
- determining the number of times an item can be renewed or the status of items
For those of you counting, you are correct that the lists do not add to 20 items. Some questions asked the same thing but in a different manor. For example, on one set of questions participants were asked to find a book by topic whereas the other set of questions asked participants to find the a book by title and author.
I am not going to tell you the changes we made as we are conducting another round of usability testing to see if the adjustments were improvements (or not). You can help us out. Look for announcements on the Library’s blog, Facebook, and Twitter.