A laptop with PDF icon.

You may have heard that, when it comes to accessibility, PDFs are trouble. But what does that actually mean? 

Until recently, I had no idea. When I taught, I’d upload scanned articles, short stories, or chapters of books to Canvas with abandon. It was so easy! So convenient! I didn’t even need to use the giant scanner on the English department copy machine – I could just scan whatever I wanted to my phone and upload it from there. That Karen Russell short story for my creative writing class? Done! That essay on gender and tragedy in Titus Andronicus? Magic!

I had no clue that, by doing this, I was gleefully circulating inaccessible content to my students. Because without intervention, nothing I scanned to PDF and shared was accessible to students who needed or preferred to use a screen reader. I was entirely unaware I could be making life harder for my students, when all I wanted to was to give them the best learning experience possible. 

But I didn’t have any students who requested an accommodation for blindness or low vision, so what was the problem? 

What I didn’t know is that screen readers – an assistive technology that converts onscreen text into synthesized speech or Braille – don’t just benefit students who are blind or have low vision. Many students can use screen readers for a variety of reasons, including students with dyslexia, ADHD, motor control difficulties, and students who simply process information better through auditory content, as opposed to visual. They can also be helpful for language learners, and even help students who prefer auditory processing with proofreading. 

So, what to do? Once again, I sat down with Julie Rummings, CTI instructional designer and accessibility specialist, to find out. 

Optical Character Recognition: The Unsung Hero of Accessible Content

An example of an inaccessible PDF, with the text highlighted in blue, signifying that it is a photo image and therefore unreadable by a screen reader
The Blue Box of Doom: Alas, this PDF I shared in my Unruly Women of the English Renaissance class was not acessible: The blue highlighting indicates the screen reader would have read this as an image, and not had access to its text.

“The biggest issue is PDFs or scans of journal articles that are not accessible to screen readers,” said Julie Rummings, CTI instructional designer and accessibility specialist. “You know how, when you click on a word in a text and the whole paragraph highlights? That’s because it’s an image. Basically, anytime you scan a book chapter to a PDF, the text doesn’t show up as text to a screen reader – it shows up as a photo.”

Unfortunately, that photo would need quite a bit of alt-text to be made accessible.

The problem lies in a little something called Optical Character Recognition (OCR) – or more accurately, a lack thereof. Ensuring a file has OCR is the most important step in making it accessible, because it’s what allows the screen reader to have access to the file. Without it, the screen reader doesn’t even know there’s a file there to read. As far as it’s concerned, it’s just a document with one giant image in it.

 “You can make scanned chapters accessible, it’s just time-consuming,” Julie said. This is because the process requires effort and a specific type of software – most commonly the paid version of Adobe. While instructors can likely get or may even already have access to this program, it takes training to learn how to use it. And while Adobe does have an accessibility checker, fixing the actual accessibility issues still requires knowledge of the accessibility guidelines. 

“If you want to do this, then that’s great because then you know forever,” Julie said.  “But it’s a lot to expect from people, especially if it’s the week before class starts.”

Eight Tips for Sharing Accessible Content – Including PDFs

Say you want to provide accessible content to your students, and make their learning journeys easier. But you’re faculty: perennially time-pressed, trying to balance research, teaching, departmental service, your personal obligations, and your sanity. And you really need to include that article or chapter in your course this semester, but you only have a PDF version. What can you do? 

You have options. 

Find an Accessible Version of Your Content:

An image of an academic article in PDF Form, with several lines of text highlighted, to demonstrate that the image is accessible and can be read by a screen reader
Finding solutions: Here’s an image of the same article in PDF form, with several lines of text highlighted, to demonstrate that the image has OCR  and can be read by a screen reader. Believe it or not, all I had to do for this was click on the image, and Adobe prompted me to “Recognize Text.” I clicked on that, and it scanned for OCR for me. Accessibility can be a rabbit hole – but don’t forget there are many ways to help you get where you need to go!
  • Search the Cornell Library System. According to Julie, some PDFs on the Library websites go through a basic accessibility review. They’re not always fully accessible, but it’s a start. If you download a PDF from the library, you can put it through Ally in Canvas to confirm that the article has OCR. It’s one small, extra step you can take that can make a world of difference to students who need the technology.
  • See if you can find the article online in a different format on a website. Linked PDFs on websites don’t count for this, but sometimes articles are available in full text on the website itself. 
  • Export the article in a different format: Some journals allow for this. If so, aim to export your article to HTML which is the most accessible – It’s the language screen readers speak and read. One caveat – if you do export to HTML, and that’s all you do, be mindful that a screen reader will not necessarily be able to tell the difference between the main text and, say, a footnote, and maybe not read the article in the correct order.

Work With the Content You Have:

  • There may be an app for that: Apps like Ally’s Alternative Formats feature in Canvas often have OCR and will automatically do their best to convert scanned text. This may get you from a 0% accessibility score to 40%, and you can remediate from there. It’s a start – and a good one!
  • Export the scanned document to a Microsoft Word document, then use Microsoft’s accessibility checker, which can be easier to format than Acrobat to remediate the document. In general, Microsoft more popular products have an accessibility checker – run that before uploading materials and do not save as a PDF
  • Don’t use Google Docs to create PDFs, which are notoriously difficult to remediate. Google Docs has many general benefits, including ease of collaboration, but it’s not accessibility-friendly when it comes to creating PDFs. First, it doesn’t have a built-in accessibility checker, which means you have to download an add-on to have a sense of your starting point.

If you save a Google Doc as a PDF it’s going to be harder to remediate,” Rummings said.  Instead, she offers, “save the Google Doc as a Word doc, fix the formatting to be accessible, and then save the Word doc as a PDF.”

Another reason to do this? When you export a Google Doc to a PDF, the export doesn’t always or reliably retain the accessibility changes you have made. So, if you’ve carefully labeled your headings in the Google doc to make sure it’s friendly for screen readers, the export may erase your work. 

Honestly? I’m a little stressed and slightly exasperated just reading that – it’s my reflexive, unfriendly reaction to unfriendly and confusing tech. But this is important: if you prefer creating and collaborating in Google Docs, don’t feel you have to give it up.  Just export your Google Doc to Microsoft Word, do your remediation there, and then export the Word Doc to a PDF. 

It feels a bit like a game of telephone, but if there’s one lesson you take from this blog, it’s that these little 

extra steps – exports, in this case – can have a big impact. 

Get Back to Basics: 

  • Keep it Simple:  While complexity of thought should always be encouraged, complexity in documents is the enemy of successful PDF remediation. Keeping your document layout simple can help you avoid future headaches. So: whenever possible, try to avoid excessive layouts and stick with a simple document. Headings, paragraphs, and a few images over a few pages are fine – it’s the long, complex documents & layouts with many complex images like graphs and charts that cause trouble for screen readers.
  • Have a PDF and are stuck with it, and starting to feel overwhelmed? Stop here, and take a breath. It’s ok to pause, do what you can as you can, and come back later. After all, “Rome wasn’t built in a day” is an adage for a reason. Showing up matters, and we’re just glad you’re here. 

Remember you can always check out CTI’s Accessibility Guide, which has a section dedicated to remediating PDFs. And CTI is here to help: you can contact us anytime for a consult, or stop by our online drop-in hours with your questions – we look forward to working with you!