The Data Unpacked

The following data was retrieved from our findings and provide specific and detailed descriptions of key findings, or discoveries, made through our research. These specific data insights show the important trends made by our participants to better predict and understand key usability issues on the Library of Congress website. A copy of the task list is provided below for easy access.

Based on the figure above, half, or 3, of our participants expressed that they would revisit the Library of Congress' website to find a historical document, while the other 3 participants expressed that they would rather use another source. 

It's clear from these findings that while some users are able to find the benefit of loc.gov's search system and database, it's retention rate for users is not as high as it could be, possibly due the complicated nature of searching for primary sources on the site. The Library of Congress offers a vast amount of primary sources in all kinds of mediums in high quality for free for anyone who frequents the site, so there is clearly value in what the website has to offer. The mixed satisfaction from our participants stems from the complex nature of searching the library's database, which for half, was not worth the effort to find a primary source. 

The website offers a library/digital catalog, in which users can only view availability and request to view items in person physically at the library, and digital collections/items, of which users can freely view online. While both of these databases are distinct, the home page search bar mixes results from both in its search results, causing our participants to click on items they thought were digitally accessible, but were only furthering the frustration in trying to find the right item. For some of our participants, this initial learning curve was worth it to find the primary source they needed, while for others, the process was deemed too complicated to be worth it.

The trend in this graph, which depicts the average time spent on each task by participants shows that users generally started off taking longer on a task, with each subsequent task taking less time up until task 4, in which the average time goes up. For the first task, the "Visitor" page was the most overwhelming, as it has many buttons and links that made it very difficult for our participants to find all of the information required from the task, an issue that could be remedied by consolidating all the visitor information onto one page, as opposed to the users having to click multiple links to find all of the information needed to plan a trip. The spread out information in regard to visiting the library thus led to some users getting lost or confused when trying to find all of the information relevant to visiting. The cluttered layout, spreading out of vital information, along with irrelevant information being added to the "visit" page shows a lack of the heuristic in regard to minimalist design, at least in part to some aspects of the library's website.

With task 2 involving the use of an online shop, users were already used to the conventions and layout of an online shop, making it quick and easy for users to navigate task 2. The use of the heuristic of matches between systems made it so that users already knew where to go to find a specific category of item, the sale items, and the cart, as the Library of Congress' shop was laid out in a manner similar to many other popular online shopping platforms, of which a user mentioned having a feel similar to Amazon. Task 3 similarly goes down in average time as users began to get familiarized with loc.gov's layout.

The trend from this graph showed that our participants look longer to complete Task 4 most likely because they expected to find the exhibition information directly from the homepage or easily accessible from the "visit" page. While both the homepage and "visit" page did contain the latest exhibition information, it was either too small or surrounded by a large amount of other information/text that made it too overwhelming for users and they, thus, ended up overlooking this - again, showing some lack of the heuristic for minimalist design. 

Task 5 also took longer for our participants to complete due to having to find an audio clip that is downloadable. While some users struggled to find an audio clip from the right time period, usually be overlooking the date published, other users struggled to figure out how to download the audio clip, as not all audio files from loc.gov can be downloaded, leading to confusion and frustration from our participants. This highlights the website's issue with maintaining consistency across its catalog, and a label for what is or isn't available to download would help with clarifying any confusion.

The data in the chart above displays the amount of errors a participant made for each task. The point of this graph is to demonstrate which tasks were the most difficult for users and which were easiest to complete. 

The data chart indicates that 1 and 4 were the most difficult for people and this is possibly due to a multitude of reasons. A likely reason for task 4 being one of the most difficult is the actual location for the "exhibitions" section on the homepage not being very apparent on the home screen and the exhibition section not being able to be searched up on the site. In this sense, loc.gov goes against conventions of other websites, in which home page search bars can usually be used to search the entire website, as the library's search bar can only be used to search either the digital catalog, digital collections, or items within digital collections. While the Library of Congress' shop does follow the heuristic of matches between systems, their homepage goes against this rule, an issue that caused frustration among many of our participants who attempted to find visitor information through the home page search bar.

As for task 1 being one the more difficult tasks, as mentioned previously, this is most likely due to the large amount of information present on the website's "visit"page, which can be overwhelming to newcomers, and since it is the first task, it's more likely for users to commit errors in the process of familiarizing themselves with how the website actually works. 

Task 5 had a good portion of medium error counts due to the nature of some audio clips having download access and others not, confusing participants. The lack of consistency between files of the same medium goes against the consistency and standards heuristic, and a label of what files can or cannot be downloaded may have helped in users being able more efficiently complete the task.

Tasks 2 and 3 went the smoothest for most users, thanks to the match between systems of the shop and the handy use of the filters available in both the search bar and search results page, making it easier for users to find images available on the website. By following the flexibility and ease of use heuristic, loc.gov made it more efficient for our participants to filter for photos specifically.

The data graph above demonstrates the amount of times a test participant was confused on each task and needed to ask for assistance or assurance from their test facilitator. 

Some conclusions that can be drawn from this chart is that task 1 seemed to be the most confusing for people since it was the first task, had the most information to find, with the most information spread out across different pages, making it the most complicated task. 

Similarly, task 5 was confusing for many participants do to the previously mentioned lack of the consistency heuristic with audio files, in which some can be downloaded and others cannot. Additionally task 3 has a drastic spike for one user, most likely due to the nature of its similarity to task 5, in which several primary sources from a very specific era had to be found. While some users were more advanced in their knowledge of using filters to more easily navigate their search, some users, such as User B, were less informed on the use of filters, making it more difficult and time-consuming to search through the digital catalog and collections.

Up to this point in the data analyzation process, along with the information presented in the post-test questionnaires, a few notable universal issues in terms of usability became apparent. One of those is that the homepage is cumbersome to navigate for anyone who has not been on the site, creating a learning curve that users must overcome. Another similarity is the fact people do not usually read or go through webpages with a large amount of information that they have to look through to get to where they want. Lastly, people now due to social media have small attention spans to read or investigate, so if they can't get to where they want to after one or two tries, they will become frustrated and leave the site/webpage. From these findings it is clear that loc.gov needs to condense its information to the most pertinent, create a more minimalist design for its homepage, and follow the conventions of other websites more closely to better improve the heuristics of it site.

When looking at the comparison of those who completed tasks and those who did not it is clear to see that task 4 and 5 were the most uncompleted.  

For task 4, the possible cause is the fact that on the website the exhibition section is pretty hard to find due to the small profile of it on the homepage, which is where pretty much all participants expected it to be. 

Then for task 5 the possible cause for having the most uncompleted is more than likely due to the amount of steps that the participant has to go through to know if they have the right audio, as well as their being only certain audio that is downloadable. This made it very confusing for the participant to find an audio recording that fit the task. 

There was a very prominent theme when looking at why these 2 tasks were the most uncompleted and it comes down to a lack of clarity from the website that makes it clear and concise the information available, or the type of file, its date, and its accessibility that makes it for a user quickly and efficiently find relevant information, primary sources, and research material. In recommendations, we go over some of the suggestions made here in the data page and expand upon them in an effort to improve the current usability of loc.gov.