Department of Writing Studies, University of Minnesota: Twin Cities
WRIT 4501: Usability and Human Factors in Technical Communication
Nick Rosencrans
March 19, 2025
This document describes a test plan for conducting a usability test of the St. Paul Police Website. St. Paul’s police website serves as the primary online resource for its residents and visitors who seek authoritative services regarding safety, personal property, community outreach, and more. As a city government resource, it serves a wide user base that features a variety of demographics, including but not limited to, different language proficiencies, disabilities or neurodivergences, and digital literacies.
Despite offering a wealth of useful information, the St. Paul Police Website features poor information architecture, which hinders navigation, efficiency, and the overall user experience. Users may struggle to locate essential information quickly, leading to frustration, especially during stressful or time-sensitive situations. The absence of a clear information hierarchy increases cognitive load, making it challenging for users to understand where to find relevant services.
The objective of this usability test is to identify and alleviate users’ cognitive loads by reorganizing the website’s content and enhancing navigational elements. Proposed solutions include the introduction of a “Quick Links” section on the homepage, featuring easily accessible links to popular, high-priority services like filing complaints or finding the police impound lot. By prioritizing the visibility of time-sensitive functions, we aim to improve overall user satisfaction and accessibility of critical information. The usability test will involve observation of five participants completing key tasks on the St. Paul police website.
Overall, the goals of usability testing include establishing and validating user performance and preference measures by addressing efficiency, ease of use, and user satisfaction.
Our usability test focuses specifically on finding the impound lot page. We chose this focus because the St. Paul Office of Technology and Communication mentioned that a sector of their target audience includes those looking for their car that was towed by the city, and that one of their goals was to optimize the content the site offered. Specifically, they asked about qualities such as efficiency, efficacy, and error tolerance. Some specific questions they asked were whether or not users could easily find what they needed, how they could leverage the design system to be more helpful for users, and what components were most confusing for users. In response, we hope to learn the following things:
Our specific research question is as follows: How does the St. Paul police website’s content and structure guide a user through finding and using popular services effectively and efficiently?
In terms of efficiency, we will address whether or not participants are able to complete tasks and we will measure the time the task-to-completion for each task. For this study, “efficiency” refers to accuracy and success within three minutes.
In terms of ease of use, we will record the number and type of issues that occur during testing sessions, and we will rate the impact of these issues according to frequency and severity. Severe issues are those that deter participants from successfully completing a task, whereas less severe issues are those that affect but do not prevent successful completion of the task. We anticipate a benchmark of 70–80% issues at the high-severity level and 20–30% issues less severe.
In terms of satisfaction, we will ask participants to rate their impressions of each task. We will also ask participants to describe their impressions and experiences in a debriefing interview. We anticipate a benchmark of medium to high user satisfaction. (In this study, we keep our benchmark as a spectrum starting with medium satisfaction to reflect the amount of issues we noticed across the St. Paul police website. For the time we have for a project of this depth and breadth, we don’t expect to realistically fix all or even most of the issues, so our benchmark will start at medium to reflect those we can’t address.)