Screen Reader Trainings

PAC delivered a three-part Screen Reader Trainings series for Clarivate, building staff understanding of screen readers as primary interfaces and translating demonstrations of JAWS, NVDA, and VoiceOver into practical insight for designing, developing, and evaluating more accessible digital experiences.

Project Description

PAC delivered a three-part screen reader training series for Clarivate to build foundational literacy around how screen readers work, how people rely on and operate them, and what that means for digital accessibility in practice. Clarivate staff needed a deeper understanding of the impact accessibility issues have on screen reader users and how to better address issues reported by them. PAC was uniquely positioned to support that work because of our commitment to centering the perspectives of people who rely on access technology.

Rather than treating screen readers as testing tools, the training framed them as primary interfaces through which many people navigate, work, communicate, and participate in digital life. That framing shaped the full series, from how core concepts were introduced to the contrasting demonstrations that made the real cost of inaccessible design concrete. Watching someone navigate at speed, hearing the reasoning behind software preferences and usage patterns, and experiencing inaccessible content from the vantage point of someone directly affected by it communicated something that documentation and testing checklists cannot.

The goal was not for attendees to become expert screen reader testers, which requires years of commitment and experience. Instead, the sessions aimed to develop genuine understanding grounded in the perspective of someone for whom a screen reader is not a test tool, but a way of interacting with the world.

Screen Reader Fundamentals

The first session established what screen readers are, how they work, who uses them and why, and what it actually feels like to interact with a computer through one. Core concepts included the difference between the screen reader cursor and system focus, linear versus element-by-element navigation, and the role of document semantics in making content navigable.

A central framework was the idea of a single point of regard. Where a sighted user can scan a page and take in its layout visually, a screen reader user builds a mental model of an interface by moving through it one element at a time. Well-structured content directly supports that model. Poorly structured content creates friction at every level, from orientation and navigation to comprehension and task completion. Live demonstrations of accessible and inaccessible websites made that distinction audible and concrete.

The session also addressed common misconceptions, including the assumption that running a screen reader is equivalent to knowing how to test with one, and that testing with a single screen reader is sufficient.

Screen Readers for Windows

The second session moved into hands-on practice with JAWS and NVDA, the two most widely used screen readers on Windows. NVDA is free, open source, and available in more than 60 languages; JAWS is a commercial product with deep roots in workplace and educational settings. While the two share many commands and behaviors, they differ in how they present web pages, handle form interactions, and integrate with applications. Those differences reinforce why testing with a single screen reader is insufficient.

The session covered desktop navigation, document reading, web browsing, and form completion, with particular attention to screen reader modes: the browse and virtual cursor modes used for reading content, and the forms and focus modes required for interacting with fields and complex widgets. A contrasting inaccessible form demonstrated the specific friction poor implementation creates, not as an abstract compliance failure, but as a concrete barrier with a real cost.

VoiceOver on macOS

The third session introduced VoiceOver, the screen reader built into macOS and the only graphical screen reader available on that platform. Where JAWS and NVDA largely encourage users to work with the operating system’s own navigation patterns, VoiceOver introduces its own cursor and command model. It also navigates through a hierarchical content structure that requires users to explicitly step into and out of grouped content.

That structural difference has direct implications for developers building macOS applications and web content accessed through Safari. The session helped participants understand not only how VoiceOver differs from Windows screen readers, but why those differences matter when designing, developing, and evaluating digital experiences.

Concluding Thoughts

The series was designed to produce more informed collaborators: people who understand what screen readers are, why they matter, and how choices made in design, content, and code play out for the people navigating through them. That understanding is not a certification or a checkbox. It is a disposition, one that changes how people ask questions, what they notice, and what they build.